Improving the Robustness of Point Convolution on k-Nearest Neighbor Neighborhoods with a Viewpoint-Invariant Coordinate Transform
Improving the Robustness of Point Convolution on k-Nearest Neighbor Neighborhoods with a Viewpoint-Invariant Coordinate Transform
复制标题
DOI：
10.1109/wacv56688.2023.00134
复制
发表时间：
2023-01
期刊：
影响因子：
--
通讯作者：
Xingyi Li;Wenxuan Wu;Xiaoli Z. Fern;Li Fuxin
中科院分区：
文献类型：
--
作者：
Xingyi Li;Wenxuan Wu;Xiaoli Z. Fern;Li Fuxin
最近，人们对不规则采样点云的卷积处理非常感兴趣。点云与栅格图像有很大的不同，点云上不能有规则的采样网格，这使得不规则邻域下的鲁棒性成为一个重要问题。特别是，k近邻（kNN）邻域对泛化提出了挑战，因为在训练和测试时间之间，邻域的位置可能非常不同。为了提高对不同邻域采样的鲁棒性，在常规三维坐标的基础上，提出了一种新的视点不变坐标变换作为点卷积权值生成函数的输入。这允许我们同时为网络提供非不变坐标、尺度不变坐标和尺度+旋转不变坐标，这样网络就可以自动学习卷积函数中包含哪些坐标。我们的经验表明，这有效地提高了SemanticKITTI和ScanNet数据集上点云卷积的性能，以及对显著测试时间下采样的鲁棒性，这可以大大改变kNN邻域中邻居的距离。实验中，在纯基于点的方法中，我们使用基于点的卷积框架KPConv在SemanticKITTI和ScanNet上实现了相当的语义分割性能，但由于使用kNN邻域而不是ϵ-ball，效率显著提高。
Recently, there is significant interest in performing convolution over irregularly sampled point clouds. Point clouds are very different from raster images, in that one cannot have a regular sampling grid on point clouds, which makes robustness under irregular neighborhoods an important issue. Especially, the k-nearest neighbor (kNN) neighborhood presents challenges for generalization because the location of the neighbors can be very different between training and testing times. In order to improve the robustness to different neighborhood samplings, this paper proposes a novel viewpoint-invariant coordinate transform as the input to the weight-generating function for point convolution, in addition to the regular 3D coordinates. This allows us to feed the network with non-invariant, scale-invariant and scale+rotation-invariant coordinates simultaneously, so that the network can learn which to include in the convolution function automatically. Empirically, we demonstrate that this effectively improves the performance of point cloud convolutions on the SemanticKITTI and ScanNet datasets, as well as the robustness to significant test-time downsampling, which can substantially change the distance of neighbors in a kNN neighborhood. Experimentally, among pure point-based approaches, we achieve comparable semantic segmentation performance with a comparable point-based convolution framework KPConv on SemanticKITTI and ScanNet, yet is significantly more efficient by virtue of using a kNN neighborhood instead of an ϵ-ball.